iDesign – Brake Disk Optimization

For a brakes design engineer, knowing the surface temperature of the brake disk is critical for predicting the brake pad wear. Surface temperature computation. is very difficult as it depends on number of variables such as material properties of disk, initial speed, final speed, disc outer diameter, inner diameter, flange thickness, car weight, and distribution of weight on front and rear axle.

iGloble has developed a cloud based solution for brake disk optimization that calculates the surface temperature of a ventilated brake disc for different input parameters mentioned above. The thickness for disk can be varied along with the mass of the disk for the most optimal disk surface temperature.

In most applications the total thickness of the brake disc is constant and the design engineer has the flexibility to change only the flange thickness of the disc. Increasing the flange thickness reduces the surface temperature, but also reduces the area of the vents available for forced convection. It increases the weight of the disc as well. This application will therefore help the design engineer decide on the most optimum flange thickness of the brake disc.

This Brake Disk Optimization application will help the design engineer to determine the brake disc surface temperature for any car, any disc size, any brake pad material and any braking severity.

In case you are interested in a demo, please contact us as

Testing new designs virtually

After a new design of a component or system has been made, lot of time and cost goes into physically testing the design to check its performance. The new design of a component/system is fitted in a car and driven over thousands of kms. Sometimes these tests go on for six months or more. Using Machine Learning, we can check the performance of a new design, virtually, by subjecting it to the same driving conditions and road conditions, the data of which has been collected previously. And once the driving and road data is available, it would take a few minutes to check the performance of the new design.

Predicting Component Failures Using Machine Learning

Components in an automobile fail over time because of wear and tear, which accelerates based on the driving behaviour, terrain, ambient temperature,etc. How do we measure that? and can it be predicted with a good amount of accuracy? If it can, this can save a good percentage of maintenance dollars in addition to lowering insurance claims and recalls cost. Machine learning and AI are driving a change … for predicting failures.

Let us take the example of a automotive brake disc and brake pad. Today, it is impossible to predict when the brake liner has completely worn off, without actually dismantling the wheel and taking physical measurements.

Huge amounts of data is generated on a continuous basis from a vehicle but very little analytics is carried on that data which means we are losing a lot of good information that can be used for real time analyses of the vehicle health and for systems such as brakes, suspension, steering, engine, etc. Machine learning can be used for analyzing the data for failure of certain components such as brakes. By processing speed, temperatures, accelerations, braking etc. data, predictive models can be created for computing disk and pad wear. Machine learning models along with AI can incorporate Driving Conditions, Driver Behaviour, Car Parameters, Road Conditions and Climatic Conditions to calculate the exact wear. This is a paradigm shift from distance travelled based maintenance (preventive) to predictive (& prognostics) based maintenance. This can be a savings of almost 5-7% of the annual maintenance spend.

In the above picture, the vehicle is predicted to have less that 15% of life left, and should be replaced immediately.

In the above picture , car 1 driver has the worst driving behaviour and hence will have to change a certain component first amongst the 5 cars. This is possible only if we run the prognostics using machine learning and AI.

Connected Cars Security

Security features for Connected Cars

Five Security checkpoints for a connected car

Cyber security has become very crucial these days for connected car environment. The rate of connected car adaptation is simply not meeting the technology changes needed in terms of modern security benchmarking.

A Jeep hack made a lot of news last year where hackers were able to completely take control of Jeep remotely miles away from their home. It’s demonstrated live by the Wire magazine here.

A modern car is typically a computer controlled machine with dozens of ECUs (Electron Control unit) inside it.

Why the ECU security matters now than before?

In a connected car environment car ECUs gets connect to a network, security becomes important when someone is able to tap into your car ECUs remotely without physically present inside the car.  Need for security obviously rises when security can be compromised remotely.

Because the car electronics was not born as internet connected data transmitting and receiving thing, the security might have been ignored in designing the ECUs which continues to be mostly designed like that today.

Back in days when the car didn’t connect to a network it was a good idea for a car’s critical systems to be built on a Controller Area Network (CAN) bus, but now the same CAN bus can be accessed through readily available ports such as an OBD2 port this can potentially act as a gateway to inside your car for a hacker/hacking device.

As modern day vehicles become more connected, they also risk becoming easier to access. Potential points of attack include:

  • Maintenance interfaces: While it is possible to attack traditional IVNs directly through a vehicle’s maintenance interface, a shift to Ethernet/IP networks would make such attacks much easier for anyone with a laptop and basic hacking skills to execute
  • Wi-Fi access points: Wi-Fi access points, if inadequately secured, offer hackers the chance to attack systems from anywhere within 10-15m of the vehicle.
  • Cellular modems: Hackers can call a car’s cellular modem, and use audio signals to launch an attack.
  • Car2x Wi-Fi: Frequently used to warn drivers approaching roadworks, Car2x Wi-Fi (based on the 802.11p standard) affords would-be attackers yet another way into a vehicle’s critical systems.

 Areas where security can be employed in a connected car:

  1. ECU and tapping into CAN buses: ECU should not accept any data coming in without a TIM (Trusted Identity module). These could be vendor specific chips which could be embedded inside the existing ECU setup.
  2. OS and Firmware : Secure OS and Firmware to modern security standards
  3. Car Applications such as infotainment : What’s needed: Application Security rules
  4. Data privacy in connected car environment : Secure access to connected car data with encrypted telematics data push to servers
  5. Access Control: The connected car should connect to a network in a lock down mode – this means the cellular element such as GSM modem should be able to send the data to a predefined white listed server IP. This can be taken care at cellular network level.

Using Internet of Things for Reducing Emissions for a Greener Earth

2015 was the hottest year in the recorded history of temperature. 2016 surpassed 2015. And this is a trend. 2017 is supposed to be very warm once again and most probably beating all the records. Delhi did not have a winter until late January and was very short. The April and May temperatures are going to be higher by another 1.5-2 degrees across India. Northeast US had one of the wettest winters in 2015 until the snow storm late January. The same trend applies to the rest of the world as well. Global climate is shifting to a point of no return because of global warming. Major reasons for this are: carbon dioxide emissions by burning fossil fuel at power plants, burning gasoline by vehicles, methane emissions from animal & agriculture such as rice paddies and finally, deforestation.

Traffic Congestion is getting worse…

Commute times have gone up in the last ten years. What use to take an hour to travel 20 km a decade back, now takes two hours for the same distance in Delhi. The average commute time in Delhi is more than 60 minutes between office and home today. The infrastructure has improved but the numbers of cars in Delhi and travelling through Delhi have grown faster and along with other modes of transportation resulting in a higher density of vehicles leading to lot more congestion and frustration.

And so is burning of fuel and higher emission

Congestion means more of stop and go traffic which leads to lower fuel efficiency and higher emission and in case the car is not healthy, this adds to the woes. A study conducted by iGloble, comprising of 50 cars driven over 250,000 kilometers across India shows fuel efficiency going down to 6 km per liter and lower as the average speed goes below 20 km per hour for a trip as shown in the picture. This means not only we are spending more money today as we are going from location A to B but also causing more wear and tear of the engine potentially leading to engine failure faster. This will result in additional carbon dioxide emissions contributing to already existing global warming situation and higher maintenance spend. For a country such as India where 80% of the fuel is imported, this is a pure wastage.

Need to focus on reducing carbon footprint…to create more efficient car!

Daily commutes have to become more efficient; the vehicles have to become healthier so that less vehicular gasoline is burnt leading to lower carbon dioxide emissions even as we work to improve the driving conditions. There is a need to understand the factors that affect the commute such as time of the travel, mode of transportation, traffic signal efficiency, and efficient routes based on time to reach the destination, fuel efficiency, vehicle health, and maintenance spend:  minimizing the overall spend by the car and fleet owners.

There are millions of devices that are part of the commute infrastructure including the vehicle, infrastructures, traffic lights, roadways, etc and are generating large amount of data every second. Each data point has information attached to it. Identifying and connecting the meaningful information to assist an action is way to go. Using IoT principles to capture real time data from within the vehicles (engine, transmission, acceleration, braking, etc.) and across the vehicles, and available infrastructure, one needs to maximize the throughput in the network with the given and understood constraints. There has to be a paradigm shift from “here and now” to forecasting the network situation with the goal of optimizing throughput with minimum spends across fuel, time, and pain. This has to be communicated to the drivers so that appropriate actions can be assisted with minimal risk. IoT connecting devices and people with process changes make a shift for a better tomorrow to lower the carbon foot print.

How OBDII – On Board Diagnostics is helping build a Safer World

85% of the road accidents happen because of human error. Out of which 36% happens because of driver distraction like taking calls & texting while driving, and fatigue because of long commutes and traffic jams. The yearly insurance claim is to the tune of $940B.  Is there a way to improve the driver and the vehicle safety?

Tracking of vehicles have been happening for some time now through GPS using the web and mobile based apps.  But that is not enough! What is needed is monitoring of the driving behavior real time and how that affects the vehicle performance and maintenance. Driving behavior includes over speeding, hard acceleration, sharp turns, excessive idling, etc. All the above mentioned affect the fuel efficiency and overall working of the vehicle. Add driver fatigue to the equation, the driver and the vehicle risk goes up. Hence, it becomes imperative to measure both the driver and vehicle risk and communicate the same to both the driver and the fleet owners.

Each car manufactured post 1996 has an OBD II (on board diagnostics) port where the OBD II compliant device can be plugged in. This device can receive data from the vehicle (GPS location, 9-axis accelerometer, and engine data) which can then be pushed to the smart phone directly using blue tooth or to the connected cars platform on the cloud using the GPRS technology., a cloud based smart connected cars platform, that receives data from the connected OBD II device real time, uses a smart analytical engine powered by artificial intelligence and neural networks to analyze the data to generate the driver risk and the vehicle risk indices. Driver risk measures the safety of the driver and the vehicle from an accident perspective and the effect the driving behavior has on the vehicle performance and the associated cost. Driver ratings are published to the owners for appropriate action. This not only improves the driver and vehicle safety but also improves the vehicle health score. The vehicle risk is a measure of the vehicle failure using the real time engine based data and driving behavior leading to excessive wear and tear. Useful life left in the vehicle is computed for predicting the time to failure. Both the driver and vehicle risk index are computed for each trip along with the fuel efficiency; another critical factor for measuring operational efficiency. Furthermore, risk indices are consolidated across the trips to understand the trending over a period of time. The ability to predict vehicle health along with recommendations at the right time for the fleets to identify the high operational cost points for potential cost savings is critical. The platform identifies the worst performing vehicles so that fleets can perform predictive maintenance and diagnostics based repairs on those vehicles for an overall health of the fleet improvement.

The Pizza theory to create Integrated Solutions

Time has come to create solutions that are cutting over from one domain to another. Data can have multiple origins and destinations. What this means is that once data is created, it can consumed by the various verticals such as manufacturing, banking and insurance verticals. This is achieved by transforming data into meaning information for analytics and decision making by that specific vertical. The key here is the type of information needed and hence the transformation that needs to be designed. The aim to use these solutions is to mitigate risks and improve operational efficiency.

“The Pizza Theory” design helps in creating those transformation solutions on the data (generated by an engine of an aircraft or bus or turbine in a hydro power plant). Data is the core of the pizza, toppings are the applications, and finally a slice is an integrated solution across the verticals.

For instance, data from an engine of the bus can be used to predict its failure and also to decide whether to repair /replace the engine. A fleet owner can decide on the fate of its fleet if the engine issue is consistent across all the buses. Same data can be used to communicate to the driver of the bus on the engine condition so that an appropriate decision can be made while in motion. Engine manufacturer can use the same data across all engines to decide if this needs a design change. And finally, insurer for the engines can measure the risk involved here. The slice of the Pizza is built on that engine data.

Impact of AI and ML on world

AI will transform the world!

Artificial Intelligence is emerging as the most powerful solution which will change the way we do things and make decisions. Like the way we are connected socially using Facebook, LinkedIn, Twitter, etc. and like the way we are connecting machines as part of IoT, we need to connect databases across the globe. Transforming that data into information is critical. Understand information is what we use for decision making and not data. Data can be used for trending and correlations. As we are talking about huge data sets here and connecting them, this is a hard problem but solvable.

Decisions are being made looking at the local information base. Connecting and running through the global information bases, AI can self-learn from local decision making and help drive decision making which are more globally optimized and faster. This applies to the verticals such as manufacturing, healthcare, energy & utilities and so on. Connecting the dots, AI can learn from each dot and improve the solution for the next dot. As more and more dots get connected, AI solution becomes smarter and we get more predictable solution.

For example, a CAT scan machine installed and being used in one country under varied environments and circumstances, can I use that information to improve the working in another country may be in similar situation or completely different situation. What I can learn from existing information and how can I improve not just the working and customer satisfaction but also the design and supply chain.